HFSA-AW: A Hybrid Fuzzy Self-adaptive Audio Watermarking

In this paper, a novel Hybrid Fuzzy Self-Adaptive digital Audio Watermarking scheme (HFSA-AW) is proposed based on local audio features. Firstly, the original audio signal is partitioned into audio frames. These audio frames are transformed into discrete wavelet transform (DWT) domain. A method for performing automatic segmentation based on features related to rhythm, timbre and harmony is presented. A proposed segmentation approach, combining audio signal characterization by statistical features and fuzzy clustering approach, is proposed. The local features of each audio frame are extracted respectively, and these features are used to train a fuzzy c-means clustering algorithm. Since fuzzy set theory is capable of performing complex nonlinear mappings between input and output spaces, it can effectively estimate the strength of a frame for each sub-band and ensure that the embedded watermark in the original audio is self-adaptive. A new watermark embedding scheme, based on fuzzy adaptive embedding strength, is used to embed watermark into the statistics average value of low frequency components. In order to evaluate the performance of the proposed audio watermarking method, subjective and objective quality tests including bit error rate (BER) and signal to noise ratio (SNR) are conducted. Experimental results show that the proposed scheme is inaudible and robust against common signal processing, including low-pass filtering, noise addition, and cropping.